Scheduled Denoising Autoencoders

Krzysztof Geras, Charles Sutton

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We present a representation learning method that learns features at multiple different levels of scale. Working within the unsupervised framework of denoising autoencoders, we observe that when the input is heavily corrupted during training, the network tends to learn coarse-grained features, whereas when the input is only slightly corrupted, the network tends to learn fine-grained features. This motivates the scheduled denoising autoencoder, which starts with a high level of noise that lowers as training progresses. We find that the resulting representation yields a significant boost on a later supervised task compared to the original input, or to a standard denoising autoencoder trained at a single noise level. After supervised fine-tuning our best model achieves the lowest ever reported error on the CIFAR-10 data set among permutation-invariant methods.
Original languageEnglish
Title of host publicationInternational Conference on Learning Representations (ICLR) 2015
Number of pages11
Publication statusPublished - 10 May 2015

Fingerprint Dive into the research topics of 'Scheduled Denoising Autoencoders'. Together they form a unique fingerprint.

Cite this